SPOREs : Organ-Specific SPORE Programs : Breast

University of North Carolina at Chapel Hill

Overall

This application proposes renewal of the UNC Breast Cancer SPORE. Originally funded in 1992, the UNC SPORE has used significant institutional investment and the distinctive SPORE elements: funding flexibility, interdisciplinary collaboration, bidirectional translational research, developmental programs, and interSPORE partners to build an outstanding program in translational breast cancer research. The UNC SPORE’s combination of population-based research, molecular genetics, breast cancer biology, health disparities research, tissue-acquiring clinical trials, database development, and expertise in bioinformatics and biostatistics was made possible with long-term SPORE support. Interaction between disciplines has resulted in substantial productivity, as measured by important findings published in excellent journals, career advancement for junior investigators, developmental research leading to grants and new SPORE projects, multiple funded interSPORE collaborations, and innovative approaches to breast cancer etiology, classification, prognosis, and therapy.

Our five projects are conceptually linked by studies of breast cancer molecular phenotypes, particularly those with the worst prognosis: basal-like tumors, luminal B tumors, and tumors that overexpress HER2. Led by basic and clinical science teams, these projects feature our continuing population science study, the Carolina Breast Cancer Study, and three translational clinical trials. Data from multiple national and international trials will also be incorporated into our analyses. The projects are entitled:

1. Carolina Breast Cancer Study: Genetic susceptibility for breast cancer subtypes in African Americans and whites
2. Breast cancer vaccine strategies for HER2 and luminal B tumors
3. Determination of breast cancer subtype sensitivities to standard chemotherapy and combination chemotherapy-biologic regimens
4. Molecular portraits of human breast cancer endothelium
5. HER4 Isoforms: Tumor suppressor action and prognostic significance

They are supported by strong infrastructure and institutional commitment from the University of North Carolina, its UNC Lineberger Comprehensive Cancer Center, and its Schools of Medicine and Public Health. SPORE investigators will have access to Cancer Center core facilities as well as SPORE funded cores. Specific to the SPORE are Genomics, Genotyping & Bioinformatics, Tissue Procurement & Analysis, Biostatistics, and Administration.

Project 1

The Carolina Breast Cancer Study (CBCS) is a comprehensive, interdisciplinary investigation into the causes of breast cancer in African American and white women. CBCS focuses on understanding how genetic and environmental factors interact to cause breast cancer. From 1993-2001, CBCS enrolled 2311 cases of in-situ and invasive breast cancer and 2022 frequency-matched controls from a defined geographic region of eastern and central North Carolina. 40% of CBCS participants are African-American. Over 70 peer-review publications have resulted. During the last SPORE funding cycle (2001-2006), we used previously collected DNA samples to conduct genotyping for polymorphisms in DNA repair genes. Genotypes at multiple genetic loci were combined to create "pathway" genotypes. We observed interactions between combinations of single nucleotide polymorphisms (SNPs) in the double strand break DNA repair pathway and radiation exposure, and combined genotypes in the nucleotide excision DNA repair pathway and smoking.

The data rich CBCS attracted biostatistics collaborators who developed statistical methods to estimate haplotypes at the individual level and to use haplotypes to evaluate gene-gene and gene-environment interactions. We propose to expand our previous investigations by conducting a comprehensive study of haplotypes in genes involved in DNA repair, damage recognition, cell cycle control and cellular proliferation, hormone biosynthesis and metabolism, and oxidative metabolism. The advent of multiplex genotyping and identification of haplotype-tagging SNPs now makes it possible to capture the principal sources of genetic variation in the candidate genes in African Americans and whites. With newly-developed statistical methods, haplotypes will be used to evaluate gene-gene as well as gene-environment interactions. Monte Carlo simulation and applied Bayesian analysis will be used to address multiple hypothesis testing. We will also genotype 100 SNPs that serve as ancestry informative markers in order to adjust for population stratification.

We used tumor blocks from CBCS cases to determine the prevalence of specific subtypes of breast cancer. The subtypes were codified using gene expression profiling. In the CBCS, the estrogen-receptor positive forms of breast cancer, Luminal A and B, were found at highest frequency in white women and older African American women, while the estrogen-receptor negative forms, including Basal-like breast cancer, were at highest frequency in younger African American women. CBCS patients with Basal-like breast cancer had lower disease-specific survival than patients with Luminal A or B. Our preliminary data suggest that combinations of genetic and environmental exposures lead to increased risk of specific subtypes of breast cancer. We will expand our investigation of genetic susceptibility to candidate genes for breast cancer subtypes, including newly-discovered polymorphisms in genes that show differential expression in breast cancer subtypes. Positive findings (including main effects and associations with breast cancer subtypes) will be repeated using DNA samples collected from a large population-based study in Norway.

Project 2

Improved outcomes for patients with breast cancer is dependent on both the earlier detection of malignant lesions when local therapy is the most effective, and a marked improvement in our current approaches to curing patients with metastatic disease. While biological response modifiers such as trastuzumab and bevicizumab have improved response rate and time to progression of patients with metastatic breast cancer, the long-term outcome remains quite poor, with fewer than 2% of patients alive 10 years after diagnosis.

Immunotherapy offers a new approach to the treatment of patients with metastatic breast cancer. Previous studies have shown that tumor cells that are resistant to chemotherapy can be effectively killed by cytotoxic T cells. Additionally, immunotherapy offers the possibility of immune memory allowing therapy in the past to remain viable into the future. Despite these benefits, immunotherapy in the treatment of patients with solid tumors outside of malignant melanoma has been disappointing. Clinical complete responses are not seen, memory does not typically persist post vaccination, and vaccines can induce T cells without effector function. Furthermore, the induction of anti-tumor responses to self-antigens may induce regulatory T cells (T_reg) and myeloid suppressor cells that block the function of our current vaccines. Thus, new approaches to tumor immunotherapy need to focus on the induction of robust immune responses that include antigen-specific CD4+ and CD8+ T cells, antibody formation and blocking the action of regulatory mechanisms.

A critical flaw in active vaccination is the lack of production of proinflammatory cytokines post vaccination critical for T cell polarization and inhibiting T_reg function.  Our group has focused on the use of viral vectors with a specific tropism for dendritic cells (DCs) that induce robust proinflammatory cytokines.  We have found that Venezuelan equine encephalitis replicon particles (VRP) encoding specific antigens induce more robust T cell response in vitro compared to the use of peptide-pulsed DCs.  Moreover these DCs are capable of generating significant quantities of proinflammatory cytokines in vitro and enhanced T cell response in vivo.

In this proposal, we will evaluate the activity of mouse DCs infected with VRP encoding HER-2/neu in breaking tolerance and inducing tumor remission in FVB-Neu transgenic mice that are tolerant to tumors expressing Neu.  In aim 1, we will investigate whether this platform leads to enhanced survival of FVB-Neu mice that have had tumors implanted prior to vaccination.  The role of type I interferons will be evaluated and whether this vaccine can induce antigen-specific CD4+ T cells assessed.  Additional experiments will determine if vaccine efficacy is enhanced by blocking the function of myeloid suppressor or T_regcells using specific chemotherapy.  In aim 2, we will determine if small molecule inhibitors of CCR5 can inhibit T_reg trafficking to the tumor site and enhance the efficacy of vaccination.  Aim 3 involves a clinical trial evaluating the efficacy of chemotherapy with VRP-Her-2/neu infected DCs in patients with metastatic breast cancer.  Finally aim 4 will evaluate whether we can generate T cell responses to the cancer-testis antigens MAGE-A1 and MAGE-A3 in Luminal B subtype tumors.  As with our previous SPORE project, our goal is to use our preclinical platform as a springboard to develop novel therapy for patients with metastatic breast cancer.

Project 3

Gene expression profiles have identified five major molecular subtypes of breast cancer (Luminal A, Luminal B, Basal-like, HER2+/ER-, Normal Breast-like) that show significant differences in survival. Mounting evidence also suggests that these subtypes vary in their responsiveness to chemotherapeutics and biologic agents. In fact, specific drug targets, like HER1 and HER2, within different subtypes suggests that logical combinations of chemotherapeutics and biologic agents may be subtype-specific, which must be empirically evaluated. Nonetheless, picking the right chemotherapeutic(s) and biologic agent combination for each subtype has yet to be experimentally or clinically determined. Therefore, we propose here to identify efficacious combination therapies for the tumor subtypes by testing at least 8 different regimens on a preclinical animal models system composed of primary human tumor xenografts, breast tumor-derived cell lines grown as xenografts, and transgenic mice that develop mammary carcinomas with known molecular subtypes. These experimental results will be augmented with human clinical trial data from several different trials that test a variety of chemotherapy and biologic therapy combinations, which should ultimately allow us to develop genomic assays that could be used to select patients for specific chemotherapy regimens, and to identify tumor subtype sensitivities that will be prospectively tested in the next generation of clinical trials.

Project 4

Virtually all drugs targeting tumor neovascularization that are being tested clinically can be classified as anti-angiogenic agents. These agents are generally cytostatic, inhibiting only new blood vessels undergoing proliferation, migration, or tube formation, and sparing established vessels. An alternative approach for targeting tumor vessels involves the use of vascular targeting agents. These agents are inherently cytotoxic, and ideally target all vessels within the tumor, both new and established. Since these compounds are cytotoxic, it is important to target only tumor endothelium, so that normal endothelium are spared. Therefore, a more detailed understanding of the molecules that are most differentially and abundantly expressed on tumor vessels is needed to facilitate the efficient clinical development of vascular targeting agents. 

We hypothesize that breast tumor endothelium are qualitatively different in their gene expression patterns from normal breast endothelium. Our objective is to obtain molecular profiles of breast tumor endothelium and compare the gene expression patterns to normal breast endothelium. We will employ our novel method for laser capture microdissection of endothelial cells from both human and mouse breast tumors and human and mouse normal breast tissue. We will extract and amplify the RNA, and proceed with genomic profiling using 44k long oligo-spotted microarrays.  We will embark on a genome-wide screen for changes in gene expression between endothelium from breast tumors and normal breast tissue, and between luminal and basal subtypes of tumors.  We will also compare gene expression profiles between mouse and human endothelium.  We will validate the differential expression of tumor endothelial markers using real time PCR, and use in situ hybridization or existing antibodies to localize transcripts or gene products as being derived from tumor endothelium.   Finally, we will see if there is a change in protein expression of tumor endothelial markers in a human clinical trial for patients with metastatic breast cancer treated with the angiogenesis inhibitor 2-methoxyestradiol and paclitaxel.  A more detailed understanding of the molecules that are most differentially and abundantly expressed on breast tumor vessels may facilitate the efficient clinical development of vascular targeting agents for breast cancer, and for developing diagnostic markers to serve as surrogate markers for response to anti-angiogenic therapy.

Project 5

Activation, overexpression, or gene amplification of EGF receptor, HER2 and HER3 are generally correlated with a poor breast cancer outcome.  In contrast, HER4 is associated with good-prognosis breast cancers and growth inhibition.  Also, in contrast to other EGF receptor family members, HER4 exhibits a unique cell biology, as it can be released from the membrane by proteolysis and the 80kDa soluble cytoplasmic tyrosine kinase is free to translocate within the cell (s80^HER4).  This HER4 cytoplasmic fragment has clearly evolved to work in the cell nucleus, as it contains three canonical nuclear localization and nuclear export sequences and, as our lab has found, functional cyclin-like D box sequences.  We have demonstrated that s80^HER4 is both necessary and sufficient to inhibit breast cancer cell line growth, including growth of HER2-overexpressing cells.  Taken together, our findings suggest that nuclear HER4 can act in a tumor suppressor-like pathway.  .  One caveat is that HER4 can be alternately spliced, to one of  two cytoplasmic domain isoforms, CYT1 and CYT2,. There is evidence that CYT1 and CYT2 may differ in their growth inhibition or proliferative signaling; our growth inhibition studies were performed with HER4 CYT1.  

Translational Objective:  To use cellular and animal models to elucidate differences in nuclear s80^HER4 CYT1 and s80^HER4 CYT2 action and to determine if HER4, HER4  nuclear localization, or specific HER4 isoforms correlate with breast cancer clinical characteristics or outcome.

Aim 1: To determine if HER4^CYT1 or s80^HER4CYT1 has similar or different signaling capabilities and biological outcomes in breast cells as compared to HER4^CYT2 and s80^HER4CYT2.  We will use human breast cancer cells in culture and xenograft to immuno-compromised mice to examine the effects of each HER4 and s80^HER4 isoform on cell growth and tumor formation. We will use the HC11 cell line, derived from mammary gland of a mid-pregnancy Balb/C mouse, which retains the ability to undergo differentiation in culture, to examine the effects of each HER4 or s80^HER4 isoform on mammary differentiation.

Aim 2: To develop quantitative RT-PCR assays (real-time) for distinguishing and measuring HER4^CYT1 and HER4^CYT2 transcripts in clinical breast cancer specimens, as well as for measuring the expression levels of HER2, BRCA1, and Wwox.  We will correlate RNA expression with clinical characteristics and outcome in breast cancer samples. 

Core 1

Historically, studies to identify cancer causing genetic alterations and interactions have been limited to single genes or proteins. Yet, alterations of biological processes involves coordinated changes in the expression of dozens or more genes. With the availability of the complete sequence of the human and mouse genomes, experimentalists can now use global approaches that simultaneously examine multitudes of genes.  In addition, the ability to collect these rich data from clinical specimens is growing rapidly.  These developments necessitate the integration of genetics and genomics with clinical medicine.  To manage and analyze the complex data sets generated, translational research must be facile with biomedical informatics, bioinformatics and computation biology. 

In response to these needs, the UNC Breast SPORE has developed Core 1, the Genomics, Genotyping and Bioinformatics shared resource.  Under the experienced leadership of the SPORE Co-PI, Dr. Charles Perou and Dr. Steve Marron, SPORE Core 1 brings together three critical services  -- high quality DNA microarray services, high-throughput genotyping, and related Bioinformatics – and focuses them on translational breast cancer research.  Four proposed SPORE projects (1, 3, 4, 5) plan to use the Core.

These three Core services are part of two UNC Lineberger Comprehensive Cancer Center shared resources – Genomics & Bioinformatics and the Mammalian High Throughput Genotyping facility.  In both these cores, the needs of Breast SPORE investigators and their studies have spurred development of methods and equipment upgrades that benefit and stimulate other Cancer Center research.  By supporting key personnel in these two Cancer Center cores, the UNC Breast SPORE is assuring priority service and performance.  Dr. Perou’s expertise and experience with these services, his leadership in the Cancer Center Genomics & Bioinformatics Core, and Dr. Earp’s dual role as Breast SPORE and Cancer Center Director mean that this SPORE Core is integrated into the Cancer Center structure and appropriately providing services to SPORE investigators.

Core 2

The SPORE Tissue Procurement and Analysis Facility (TPA) provides centralized, quality controlled, quality assured procurement, processing, analysis, storage and distribution of normal and malignant breast tissue from UNC Hospitals. Whenever possible, corresponding serum and plasma are also obtained. All procured specimens are reviewed by the Facility Pathologist to ensure that representative tissue is banked and distributed.  Our tissue procurement activities have expanded to include Rex Hospital, a large hospital in Raleigh, N.C., new to the UNC Health Care System, dramatically increasing our access to breast specimens. 

The TPA facility procures and distributes human specimens from patients who have consented to having their specimens used in basic, translational, population and clinical cancer research. The facility also provides analysis services, including tissue sectioning for frozen and paraffin embedded specimens, preparation of tissue microarrays (TMAs), immunohistochemistry and fluorescence in situ assays, technology development  and access to state-of-the-art tissue-based technology, including automated equipment and laser-capture microdissection.  Expert personnel provide consultation and training to faculty and staff in these applications.  The facility also serves as a central clearing house for the Carolina Breast Cancer Study—banking and distributing unstained sections for multiple assays and centralized repository for the InterSPORE ISPY trial (Intergroup Neoadjuvant study: SPORE/CALGB/ ACRIN).  Customized databases serve to monitor and survey all specimens and associated data, providing a coordinated system of quality control, sample tracking and distribution of specimens to appropriate investigators. 

The TPA has implemented policies and procedures to support these services and address technical, medical and legal issues, as well as protection of patient privacy and confidentiality. These policies developed in collaboration with pathology, surgery, the clinical protocol office and integrated with clinical and basic research policies allow a uniform, streamlined coordinated process of tissue procurement, handling, process, triage and analysis.  The facility meets with each investigator to customize needs. The productivity and services provided by the TPA continues, as evidenced by numerous publications from user investigators and continued support of each of the SPORE projects and developmental studies.

Core 3

The Biostatistics Core Facility provides statistical consultation, analysis and data management support to UNC Breast Cancer SPORE investigators.  The SPORE Core is extended by the UNC Lineberger Comprehensive Cancer Center’s Biostatistics and Data Management Core. The SPORE Core offers study design, sampling, data management, analysis, and manuscript preparation.  Data management activities include: database design, implementation, documentation, maintenance and user support.  Statistical support includes: implementation and programming of specific sampling schemes, assistance in clinical protocol preparation, and choice of proper design and statistical analysis methods and development of special software programs or macros for handling non-standard situations. 
The SPORE’s Biostatistics Core benefits from substantial investments in biostatistics and data management made by the UNC Lineberger Comprehensive Cancer Center.  The Cancer Center’s Biostatistics and Data Management Core was recently renewed with an annual budget of $370,000 in CCSG funds that includes partial salary support for five faculty members, four full time biostatisticians, and two graduate research assistants.  The Cancer Center Core is led by Dr. Joseph Ibrahim, Alumni Distinguished Professor of Biostatistics in UNC’s School of Public Health.  In addition to providing statistical consultation for the SPORE, Dr. Ibrahim will work with Dr. Qaqish to ensure all SPORE projects are handled by the appropriate faculty and staff with rapid response.

This Breast SPORE Biostatistics Core resource is much more than a service core.  It is a conduit to superb biostatisticians and programmers who are intimately involved with development, planning, data acquisition, algorithms, and analysis of SPORE studies.  Our SPORE’s concentration on large data sets, such as genotyping, microarray, and somatic mutation in groups of patients and controls exceeding 4,000 has provided an outstanding venue for the development of biostatistical methods.  In our SPORE, the biostatisticians are an integral part of the experimental process. 

The budget requested for this core will provide a small amount of salary for time dedicated to the SPORE and future areas for development.  With top-ranked Departments of Biostatistics and Statistics on the UNC campus, both the SPORE and Cancer Center Biostatistics Cores have a rich pool of faculty with whom to work and direct clinicians, geneticists and biologists. Our biostatisticians/statisticians, in turn, gain a superb opportunity to analyze large, important data sets.

Core 4

Under the direction of Drs. Shelton Earp and Charles Perou, SPORE PI and Co-PI, the Administration Core continues to support the UNC Breast Cancer SPORE's overall scientific/ translational goals providing leadership and day-to-day administration.  The Core organizes:  intra and inter-SPORE interactions; administrative/scientific oversight of all research projects, cores, and developmental programs; interaction with NCI and the SPORE program; and, planning and evaluation activities.  This Core also monitors SPORE expenditures and addresses grant management issues.  Like the SPORE itself, the Administration Core blends experienced and stable leadership with new expertise, energy, and perspective.  Dr. Earp is Cancer Center Director.  This Administration Core links the Breast SPORE to the UNC Lineberger Comprehensive Cancer Center and its other large translational efforts (e.g. GI SPORE, DoD Prostate Cancer Center of Excellence, Center of Cancer Nanotechnology Excellence), as well as to the University, School of Medicine, and Hospital administration.  Dr. Perou has moved into an active co-PI role during the planning of this renewal.  We believe that this strengthens day-to-day operational leadership by providing a combined 55% effort (Dr. Earp 25%, Dr. Perou 30%) for two leaders with complementary talents and roles.  This broad leadership collaboration brings expertise in genetics, bioinformatics and tissue-based translational technology while continuing senior mentorship and institutional leadership.

The UNC Lineberger is an NCI-designated Comprehensive Cancer Center and the largest research Center on the UNC campus.  The Center’s administrative team, which continues to support the SPORE, has extensive experience in directing large, interdisciplinary grants and directly handles more than $60 million dollars per year in state, overhead, private philanthropy, and grant operational funds.  Administrative Core members meet weekly to review administrative, financial, developmental, and program issues.  Topics include:  budgets, space, special requests, project and core updates, developmental project progress, upcoming seminars, consultant/advisor visits, potential breast cancer initiatives, and trouble shooting. 

The SPORE monitors progress and plans future activities through an annual review by the Executive Committee, patient advocates, and the Canter Center’s External Advisors. 

The page is maintained by: Website Contact